Power semiconductor devices, such as those fabricated from SiC (silicon carbide), may be designed to operate at very high operating temperatures (e.g., greater than 250° C.). Such power semiconductor devices may be bonded to a cooling device, such as a heat sink or a liquid cooling assembly, for example. The cooling device removes heat from the power semiconductor device to ensure that it operates at a temperature that is below its maximum operating temperature. The bonding layer that bonds the power semiconductor device to the cooling device must be able to withstand the high operating temperatures of the power semiconductor device.
Transient liquid phase (“TLP”) sintering (“TLPS”), or diffusion bonding, or soldering are methods of high temperature bonding that may be used to bond one substrate to another (e.g., a power semiconductor to a cooling device). For example, TLP bonding results in a bond layer having a high temperature melting point. A typical TLP bond consists of two different material compounds: a metallic layer and an intermetallic layer or alloy. Generally, the intermetallic layer having a high-remelting temperature is formed during an initial melting phase wherein a low melting temperature material, such as tin, diffuses into high melting temperature materials, such as copper, silver, or nickel. Conventional methods for assessing TLP sintered interconnect microstructures have utilized homogeneous virtual models, while such microstructures are generally heterogeneous in reality. However, computational power restrictions have prevented the use of heterogeneous virtual models, which would provide for a more accurate simulation of a formed real world bond and its associated mechanical, thermal, and electrical properties.
Accordingly, a need exists for alternative methods for assessing a strengthened bonding layer between a pair of substrates and formed from a high temperature bonding, such as virtual models utilizing heterogeneous models of a final sintered interconnect microstructure that do not require excessive computational power to operate.